Leading modellers who predict that swine flu — influenza A(H1N1) — could eventually infect a third of the world's population have received a mixed response from other scientists.
Researchers from Europe and Mexico published a study in Science this week (11 May) stating that the virus has full pandemic potential and could kill four in every 1,000 infected people, though they did not estimate the total death toll.
This makes the virus about as dangerous as the strain responsible for the 1957 pandemic that killed two million people but far less deadly than the 1918 Spanish flu pandemic.
"This virus ... is likely to spread around the world in the next six to nine months and when it does so it will affect about one-third of the world's population," lead author Neil Ferguson, from the UK-based Imperial College London, told BBC Radio 4's Today programme.
But influenza epidemiologist Lone Simonsen of the US-based George Washington University describes the Imperial study as the worst case scenario.
She says there is an approximately ten-fold uncertainty in the estimate of the number of deaths and a one- to two-fold uncertainty in the number of infected cases in the study.
"I think it's a completely difficult estimate to make at this point," she says.
However, new studies of past flu pandemics are beginning to reveal some commonalities about how influenza viruses evolve and spread, she says. One common feature is that the viruses cause disease in waves and often become more virulent months after they first appear.
Alessandro Vespignani of the US-based Indiana University, who is also modelling the spread of influenza A(H1N1), says that the Imperial College results are compatible with his own. "We all agree that in the long run this epidemic may affect one in three people." This is likely to happen within 12–18 months, he says.
But he cautions that there are still many unknowns: "I find it a bit of a stretch to talk about what will happen in one year."
Vespignani's model shows that Africa will be largely unaffected by the virus in the short term — and developing regions in general can expect mild outbreaks of the disease.
He predicts many more countries will be affected by the end of May this year, but the number of infected people is likely to be lower than anticipated.
Brazil will have a maximum of 300 reported cases and Colombia will see around 200 cases by the end of May, while Costa Rica will have 1,000 and Venezuela 600. Meanwhile, China and India will have around 40 cases each.
Severe gaps in information about the virus mean that modellers' predictions could change dramatically and quickly, he adds.
The WHO has confirmed that there are 5,758 laboratory-confirmed swine flu cases in 33 countries, with 61 deaths so far.
Link to full paper in Science [730kB]
Science doi 10.1126/science.1176062 (2009)